1. Field of the Invention
The present invention relates to an orthogonal frequency division multiplex and code division multiplex (OFDM-CDM) transmission system, and a transmission device (modulator) and a reception device (demodulator) for the system, and more specifically to an apparatus and a method for realizing the communications between a base station and a mobile station in the cellular phone system or the mobile phone communications system.
2. Description of the Related Art
Conventionally, an orthogonal frequency division multiplex (hereinafter referred to as an OFDM (orthogonal frequency division multiplex)) transmission system in a terrestrial digital television, etc. In the OFDM transmission system, data is transmitted using a plurality of subcarriers having different frequencies. Practically, in this system, a number of subcarrier orthogonal to one another are modulated in transmission data, and the subcarriers are frequency division multiplexed and transmitted. In the OFDM transmission system, even if high-speed data transmission is performed, the transmission rate can be lowered, that is to say, the transmission rate can be reduced for each subcarrier. Therefore, the influence of multipath interference can be reduced. The OFDM transmission system is described in, for example, “Overview of Multicarrier CDMA” (Hara et al., IEEE Communication Magazine, December 1997, pp 126-133), or “WIDEBAND WIRELESS DIGITAL COMMUNICATIONS”, A. F. Molisch Prentice Hall PTR, 2001, ISBN:0-13-022333-6).
FIG. 1 shows the configuration of an existing transmission device for use in the OFDM transmission system. In this system, it is assumed that the transmission device multiplexes signal series Si and signal series Sj and outputs them. It is also assumed that the symbol period of the signal series Si and the signal series Sj is “T”. Additionally, the signal series Si and the signal series Sj can be, for example, signals to be transmitted to different mobile stations. Otherwise, data to be transmitted to a plurality of mobile stations can be time-division-multiplexed in the signal series Si.
Each piece of the symbol information of the signal series Si is input in parallel to the m respective input terminals provided for a spread modulator 1. That is, the same symbol information is input in parallel in each symbol period T to each input terminal of the spread modulator 1. Then, the spread modulator 1 modulates the input symbol information using a spreading code Ci assigned to the signal series Si in advance, and outputs resultant spreading signals of m bits. The spreading code Ci is configured by “Ci(1)” through “Ci(m)”, and is one element in the orthogonal code series.
A subcarrier modulator 2 generates m subcarriers having different angular frequencies ω1 ˜ωm. The angular frequency interval Δω of ω1, ω2, ω3, . . . , ωm is a predetermined value defined by a reciprocal of the symbol period T, and is represented by the following equation.Δω=2πΔf=2π/T 
The subcarrier modulator 2 modulates m subcarriers using the spreading signal output from the spread modulator 1. Practically, for example, a subcarrier having the angular frequency ω1 is modulated according to the symbol information multiplied by “Ci(1)”, and a subcarrier having the angular frequency ωm is modulated according to the symbol information multiplied by “Ci(m)”. These subcarriers are combined by an adder 3.
As shown in FIG. 2, a guard interval insert unit 4 inserts a guard interval fixedly determined in advance to a composite signal output from the adder 3 for each symbol. The guard interval is inserted to remove the multipath influence of a wireless transmission line. FIG. 2 shows the state of the guard interval inserted into each subcarrier. Practically, these subcarriers are combined.
An adder 5 adds up a composite signal corresponding to the signal series Si obtained as described above and a composite signal corresponding to the signal series Sj obtained in the similar process. A guard interval is inserted into each of the composite signal corresponding to the signal series Si and the composite signal corresponding to the signal series Sj. The output of the adder 5 is converted into a predetermined high-frequency signal by a transmitter 6, and then transmitted through an antenna 7.
FIG. 3 shows the configuration of an existing reception device for use in the OFDM transmission system. It is assumed that the reception device receives the signal series Si from a radio signal transmitted from the transmission device shown in FIG. 1. In FIG. 3, the frequency synchronizing capability, the timing synchronizing capability, etc. required to receive a signal are omitted.
The signal received by an antenna 11 is converted by a receiver 12 into a baseband signal Srx, and then converted into m received signal series by a subcarrier demodulator 13. Then, a guard interval deletion unit 14 deletes the guard interval from each received signal series. For the inverse spreading of each received signal series, a spread demodulator 15 multiplies each received signal series by the spreading code Ci which is the same as the spreading code used in the transmission device. Then, by adding each signal output from the spread demodulator 15 using an adder 16, the signal series Si is regenerated.
Between the transmission device and the reception device with the above-mentioned configuration, the signal series Si is transmitted using a plurality of subcarriers f1˜fm shown in FIG. 2. The signal series Si is configured by the symbol information having the value of “+1” or “−1”. That is, the signal series Si is changed into “+1” or “−1” in the symbol period T. Furthermore, a signal transmitted using each of the subcarriers f1 ˜fm is spread-modulated by the spreading code Ci (Ci(1), Ci(2), . . . Ci(m) respectively). In FIG. 2, the bit marked with “*” indicates that the output of spread-modulation is inverted (conjugate) output because the signal series Si is “−1”.
As described above, a guard interval is inserted into a transmitted signal for each symbol. In the example shown in FIG. 2, the guard interval Tg is inserted in the symbol period T. Therefore, the inverse-spreading/demodulating process is performed in a section (section Ts) obtained by removing the guard interval Tg for each subcarrier in the reception device. Thus, multipath interference (interference generated by a delay wave) can be removed in the reception device.
Since the guard interval Tg is inserted to remove the multipath interference, it is necessary to set the length longer than the maximum transmission delay difference in the transmission link. The “maximum transmission delay difference” refers to the difference between the minimum propagation time and the maximum propagation time obtained when a signal is transmitted through a plurality of paths from the transmission device to the reception device. For example, in FIG. 4, assume that the signal transmitted through a path 1 first reaches the reception device, and the signal transmitted through a path 3 last reaches the reception device, then the maximum transmission delay difference is represented by the difference between the propagation time of the path 3 and the propagation time of the path 1.
However, in the cellular communications system, radio signal is normally transmitted from one base station to a plurality of mobile stations in a service area. The maximum transmission delay difference of a signal transmitted from a base station to a mobile station becomes larger as the distance between them increases. Assuming that the multipath interference is to be removed from all mobile stations in the service area, it is necessary to remove the multipath interference in the mobile station located farthest from the base station. Therefore, if the multipath interference is to be removed from all mobile stations in the service area, then it is necessary to set the guard interval Tg larger than a maximum transmission delay difference in a case where a signal is transmitted to a mobile station located farthest from the base station. For example, in the example shown in FIG. 5, it is necessary to set the guard interval Tg larger than the maximum transmission delay difference obtained when a signal is transmitted from the base station to the mobile station MS3.
However, if the difference in guard interval is determined as described above, the guard interval is unnecessarily long when a signal is transmitted to a mobile station (the mobile station MS1 in FIG. 5) located near the base station. In the meantime, the power of the signal in a guard interval is not used when a signal series is regenerated in a reception device. Therefore, if a guard interval is determined as described above, power is wasted when a signal is transmitted to a mobile station. As a result, the total transmission capacity of the entire communications system is reduced.